Estimation of vehicle movement in urban areas from thermal video sequences

E. Michaelsen, U. Stilla

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

In this contribution we describe a method to assess the activity of vehicles based on airborne image sequences taken by an infrared camera. At the resolution of approximately one-meter vehicles appear as elongated spots. Active vehicles often appear as a configuration of a dark and a bright spot close to each other, where the latter results from the warm motor hood. For the estimation of vehicle movement a precise measurement of the sensor movement is required. The sensor movement is inferred from image sequences. A robust estimation avoids outliers in the set of point correspondences. A special selection strategy prefers samples that promise estimation precision.

Original languageEnglish
Title of host publication2nd GRSS/ISPRS Joint Workshop on Remote Sensing and Data Fusion over Urban Areas, URBAN 2003
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages105-109
Number of pages5
ISBN (Electronic)0780377192, 9780780377196
DOIs
StatePublished - 2003
Externally publishedYes
Event2nd GRSS/ISPRS Joint Workshop on Remote Sensing and Data Fusion over Urban Areas, URBAN 2003 - Berlin, Germany
Duration: 22 May 200323 May 2003

Publication series

Name2nd GRSS/ISPRS Joint Workshop on Remote Sensing and Data Fusion over Urban Areas, URBAN 2003

Conference

Conference2nd GRSS/ISPRS Joint Workshop on Remote Sensing and Data Fusion over Urban Areas, URBAN 2003
Country/TerritoryGermany
CityBerlin
Period22/05/0323/05/03

Keywords

  • Infrared surveillance
  • Machine vision
  • Object recognition
  • Urban areas
  • Vehicles

Fingerprint

Dive into the research topics of 'Estimation of vehicle movement in urban areas from thermal video sequences'. Together they form a unique fingerprint.

Cite this